Nuclear-fusion lab achieves ‘ignition’: what does it mean? | Hacker News Reader

Nuclear-fusion lab achieves ‘ignition’: what does it mean?

nature.com

169 pointsSiempreViernes121 comments

So, they generated 3.15 MJ energy from the target, at a cost of an input of 2.05 MJ laser light. Pretty exciting.

However those lasers aren't very efficient, generating the 2.05 MJ required more than 300 MJ of electricity.

What is the path forward... Can anyone elaborate? Do we make the lasers 100x more efficient? Will hitting a target with a 100x mass yield 315 MJ?

Edit: thanks everyone for the clarification. Really helps me putting the numbers in context.

Apparently the semiconductor industry already has lasers that are 20x more efficient than what was used in this experiment.

Interesting, thanks! So the low laser efficiency isn't really important in interpreting the importance of this news.

The semiconductor lasers cannot produce pulses short enough for such applications.

They can be used only as the pumping sources for the solid-state lasers or fiber lasers that can produce the short pulses.

So the efficiencies of two lasers must be multiplied. I doubt that there is any chance to obtain an electrical to light conversion efficiency above 10%.

So the energy generated by the fusion reaction should be much greater than 10 times the light energy, to be able to produce excess energy, even with lasers 10 times more efficient than now.

Even if semiconductor lasers could put out the power required (they can't...), a 20x improvement would still mean the efficiency is only around 20%, so you'd still be putting in 10 MJ to get out 2 MJ...

Yes, it's better, but the argument still holds

I don't get it. Wouldn't that mean it required 302.05MJ to get 3.15MJ?

The purpose of NIF is to investigate the practicality of inertial confinement, not "be a production fusion reactor". They don't use bleeding edge laser tech, nor are they really built to generate significant amounts of power. For context, modern lasers are ~20x more efficient than the ones that NIF is using.

Reporting based on the energy put into the reaction itself (the lasers) reflects far better the feasibility of the confinement tech than spending an inordinate amount of budget on an already incomprehensibly expensive science experiment to try and optimize every last edge condition.

Aha that makes sense, it's a jumpstart basically?

proof of concept, yeah.

The purpose of NIF is primarily to study the effects of age on our nuclear weapons stockpile by simulating the x-ray blast of the primary nuclear device on the secondary fusion stage within an h-bomb in a controlled environment.

Then we can tell how the hydrogen stage will be affected by age, and probably how we can optimize the contents and casing of that to be smaller and last longer.

Any benefits to the future of power generation is a secondary goal.

Yes.

In nuclear fusion, "ignition" means that you get more energy out than the lasers impart. This is the first time anyone has pulled that off in decades of trying, which is a big deal.

But practical nuclear fusion would of course require that you get out more energy _from the entire system_ than you put in, and getting there will require even more problems to be solved. But now at least one obstacle has been overcome!

> "ignition" means that you get more energy out than the lasers impart. This is the first time anyone has pulled that off in decades of trying, which is a big deal.

The term is refreshingly intuitive. For decades we've been striking stone and flint. Sparks. We just got a bit of kindling going. It's no bonfire. But it's a big step forward.

>This is the first time anyone has pulled that off in decades of trying, which is a big deal.

Are you sure of that? This is from 2014: https://arstechnica.com/science/2014/02/giant-leap-for-nucle...

Quoting from the article you linked to:

    Hurricane’s current output, although more than the hydrogen fuel put into the reaction, hasn’t yet reached the stated goal to achieve “ignition," where nuclear fusion generates as much energy as the lasers supply. At that point it might be possible to make a sustainable power plant based on the technology.

from your article:

> Hurricane’s current output, although more than the hydrogen fuel put into the reaction, hasn’t yet reached the stated goal to achieve “ignition," where nuclear fusion generates as much energy as the lasers supply.

How do fusion bombs achieve "ignition"? It's not with lasers, right?

Also, couldn't the net energy from the ignition be used to continue the reaction, so you only need the lasers for startup?

Sorry for the newb questions, just curious.

Caveat: This comment is derived from publicly available info. An H-bomb is a two part bomb. The primary bomb is a implosion plutonium bomb similar in make to the nagasaki/trinity bomb. It is theorized the Xrays from that bomb are focused on a secondary cylinder of uranium-238 that has a core of hydrogen and/or lithium. The Xrays ablate this casing at such intensity that the inward force (the opposite force of the ablation) is strong enough to propell the casing inwards at 300 km/second. This squeezes the hydrogen/lithium and causes fusion which releases a tone of energy, but more importantly a huge amount of neutrons. These neutrons cause the 238 in the casing to fission releasing an even greater amount of energy. A Neutron bomb for example replaces some/all of the 238 with lead, so the neutrons escape without causing more fision. Again, this is all speculation from scientists outside the various labs, and using publicly available information.

For power generation, since the energy from any fusion reaction is primarily emitted as heat and neutrons. The neutrons are difficult to utilize and is basically wasted energy, but the heat would just heat water, turning it into steam, then turning a turbine.

With a fission bomb.

Don’t have a good explanation for your second question, although I think the short answer is “not with the way the Livermore experiment was set up.”

Kind of.. It's like saying you used a 1% efficient 100HP pump to start a water wheel that eventually was outputting 1.5HP of power. We know how to make more efficient pumps, so it's not terribly interesting that it took 100HP of power to generate the 1.5HP -- the fact that only 1HP of output power was actually used to start the 1.5HP reaction is the big deal.

Yes, but the name "ignition" is very useful.

The idea, ultimately, is that like with a wood fire, the energy from the fuel "burning" is what starts the next bit of fuel "burning", and then as it runs on it's own as long as you keep giving it more fuel, you collect the excess heat by boiling water.

So, similar to a wood fire, you might need to use a blowtorch to get it started, and run at a net negative of energy, but the exciting thing is that there was a little flame... that means we can probably make a roaring fire out of it.

No it doesn't. If you throw some fuel in their general direction, oxygen fires are self-sustaining with relatively little effort. It takes more effort to stop a large oxygen fire than to make one.

Fusion is the opposite. Fusion's natural state is Not Fusing, so in ICF you have to keep compressing and heating the fuel. Using equipment with optically tight tolerances and epic pulsed energy densities. Which are somehow maintained reliably for long periods. In spite of significant debris and huge temperature swings.

Fire's natural state is burning only at standard temperature and pressure (or greater). If you look at the "average" state of all space in the universe, fire is not stable.

Scaling the NIF's approach (laser inertial confinement) isn't about more reaction mass per pulse, it's about more frequent pulses, and is probably not going to be practical or cost effective in the foreseeable future. The fuel pellets are tiny (~2mm), their container is only about 1cm and must be nearly perfectly spherical. A continuous power plant would consume about half a million of them per day (6 per second), all of which need to be shot with perfect timing, precision, and reliability to the exact center of the laser sphere. My guess is magnetic systems are much more likely to ever be used in power generation.

This feels important. We already know that fusion "works" (proof: look up). But it's not possible to recreate the conditions of a star on Earth. So if this method is also impractical, why would I care that it "works"?

Scientifically it's potentially very interesting: close up observation of what's going on in the center of a star and all that. I'd ignore all "the future of free clean energy is upon us at last!" hype. That's the lab and the government officials trying to drum up publicity that can then be milked for financial/political capital (it costs many billions to build and run a facility like this), coupled with lazy news outlets trying to milk sensationalist headlines for eyeballs.

This seems far more credible than the "This means fusion in ten years!" hot takes, which are - frankly - completely implausible, whether they're coming from physicists or journalists.

Weapons research might be a motivator too.

I would be astounded if this can be commercialised. Not because I'm cynical for the sake of it, but because I cannot in good faith see a credible engineering path which might make that possible without spectacular breakthroughs in a number of disciplines.

MCF seems more likely to get somewhere. But my best guess is that advances in renewables will have made fusion redundant within ten years. There is far more scope for incremental and affordable improvements in collection, distribution, and storage than there is for the Big Science breakthroughs required to build one commercially viable ICF plant.

Funding for NIF is 100% about weapons, it's never been about energy. After '97 (when Clinton signed the CTBT and the US agreed to stop testing new weapons) the national labs were in a panic until some clever DOE wonk invented the concept of "stockpile stewardship" and they went back to business more or less as usual.

It proves that net positive fusion is achievable here with known physics and technology (outside bombs which are useless).

We had never done that before.

This will probably unlock more funding. In the end a different approach is probably more likely to lead to a power plant unless truly gigantic advancements in laser efficiency can be achieved.

If you create an impractical mechanism to prove out a scientific problem, you know it's then worth trying to fix the impractical mechanism. Maybe they start working on a way to direct the pellets into the correct path using magnets or something. But if you don't even know the scientific part works it's not worth going after the magnet solution.

I think OP's point is that the sun already presents itself as an impractical mechanism

It's easy to argue that an impractical mechanism we know how to make is easier to modify and refine into a practical mechanism compared to an existing impractical mechanism we don't know how to replicate.

Let’s say there are 20 different system used to make this ( laser powering, energy collections, target energy density etc etc ) , we only need to make 10 of those system 200 percent efficient to increase overall efficiency to 1000 times , within range of usefulness, I think we will be achieve this.

This is the third thread today with pithy "this means nothing takes". If you're thinking of posting a one liner on the subject you should probably stop.

Let's save some time out of discussion now:

1. The lasers at NIF are not designed to be efficient, and they are very old. Modern semiconductor lasers are 20x more efficient. The 300MJ wall plug conversation is stupid, stop pointing it out.

2. Dropping crystals and hitting them with lasers is not hard, we do it in semiconductors for EUV light source generation all the time. In fact EUV light sources do it at 100khz or more.

3. Power extraction from a bundle of hot material is not a particularly unknown problem. It's in fact how every power plant on the planet works.

Make no mistake there are a lot of engineering problems to solve. The "droplets" are enormously expensive, the "firing" itself creates debries which need cleared from the chamber, dt supply is not something that's readily available in large quantities, blah blah. I'd love interesting conversation about what stands between ICF and a real energy plant, not hot takes.

> Power extraction from a bundle of hot material is not a particularly unknown problem. It's in fact how every power plant on the planet works.

River power spinning a generator? Wind farm power? Are those counterexamples?

Well, these aren't "bundles" per-se, but as I understand it, both are relying on the Earth-as-a-system converting heat from the sun falling on particular areas into large-scale flows, which we humans can then hook into.

GP was likely referring to power plants that burn things in order to heat up water to make steam in order to rotate an iron core encircled by copper wire. Using kinetic energy like wind or flowing water is a nice shortcut to the rotating iron core. Even nicer: Cut out the rotating iron core and directly produce voltage via photons hitting a semiconductor.

I share GP‘s sentiment though that, in the end, there are remarkably few concepts at the bottom layer of electricity production.

(There are other means of course, but they are not in use for power generation in amounts that can drive even one household, AFAIK)

> 3. Power extraction from a bundle of hot material is not a particularly unknown problem. It's in fact how every power plant on the planet works.

Coal and nuclear are uncompetitive simply from the cost of the steam side. Today you can just about give a steam plant free energy and it still makes a loss.

Solar or wind does not have this limitation. CCGT gas plants gets around it by having a turbine giving raw mechanical power and then utilizing the same awful steam side to get the last percentage points of efficiency at a much smaller required scale.

Unless you can step around the steam turbine I do not see this becoming anything outside of incredibly small niches.

Neither coal nor nuclear are uncompetitive.

Compare newly built renewables with all costs included to the marginal cost of paid of coal and nuclear. [1] Today it is often better to simply shut down your existing nuclear plant and build a renewable instead. For a more comprehensive look see [2].

[1]: https://www.lazard.com/media/451885/grphx_lcoe-07.png

[2]: https://www.lazard.com/perspective/levelized-cost-of-energy-...

Looks like you'll pay the same price if you shut down an existing nuclear plant for wind/solar, unless you can bag some subsidies.

>...Today it is often better to simply shut down your existing nuclear plant and build a renewable instead.

That is ignoring capacity factor.

From your link we see that the most expensive power generation of all is residential rooftop solar. If cost is the concern, why support the most expensive option that also has a low capacity factor?

This always makes me laugh.

What other power source is accessible to the home owner?

If we are going nuclear and it’s so efficient and cost effective, why is it always subsidised so heavily? Its full cost has never been paid anywhere, as storage etc stretch off way into the future.

It seems eternally close to being cheap and reliable, but we keep screwing it up.

Meanwhile what’s left of European nuclear industry is keeping the lights on

You mean half the French nuclear fleet which is out of commission causing much of the current problems?

France has gone from the largest electricity exporter in Europe to an importer.

Which is due to the irrational fear caused by Harrisburg. The required production of new plants was halted, and now we are stuck with these aging plants.

I reckon solar is competitive in the day time without storage or peaker plants to support low generation time. I’d love to see the analysis of solar with the built in assumption of burdening the hourly cost with the infrastructure required to generate the required power at 10pm.

The generation during the day will not be burdened by that infrastructure. The night time generation will simply be day time cost + infrastructure.

Theoretically, you can bet that your traditional power plant only runs at night. But now your capacity factor is less than half the available hours for the same capital investment and it looks even bleaker.

This is where I think it probably should be. Solar w/o another generation or storage source doesn't accurately represent the full picture of what is really required to support the a whole day of electricity. So the combined total cost is probably what should be shown to represent the costs of a full day cycle of solar, or a rainy day.

My cynical take is a lot of people are subconsciously unhappy with how BS and mundane their "tech" jobs are. They chose money+cool, only it turns out the 'cool' part was a lie to get them to churn out as much useless BS work as possible. So they get through the day by fooling themselves: slap a cool tech name on your product, make trendy graphics ( https://traefik.io https://github.com ), pretend you're doing some sci fi shit. Then any real tech happening in the world exposes your BS so you need it to be framed as a disappointment.

I'm 42 and there have been "breakthroughs" in fusion announced at least once per decade of my life. There's a very real chance that this breakthrough means the first fusion power plant will open after I'm dead, as none of the men in my family have reached 80.

So, the issue is that the community here has grown old, cynical and unable to appreciate anything that might only help the next generations?

No, the community here is skeptical of science reporting that seems too good to be true and progress in fusion energy production fits the mould. I’m not an expert on fusion or the physics, so the best I can do is agree with this sentiment and note details from skeptical comments.

Hey, if we’re wrong life will change for the better for billions of people. I like being surprised with good news more than disappointment.

certainly the need to exaggerate the news has crept through all facets of media in this annoying "attention is our commodity" era. even the most "boring" outlet has its more subtle versions of the cringey YT "wow face" thumbnail.

but just because fusion energy will not be responsible for warming your frozen pizza next week does not mean we need to belittle a major milestone like this one. sure some damn reporter has a quota to hit and metrics to game. it just means we need to dedicate a bit more effort to read the article, read between the lines, and apply critical thinking.

this is the dream technology of the far future we're talking about. it doesnt even need to happen for 1000 years. if we nail it we are set for Billions of years. but obviously people are not accustomed or incentivized to think about things on that sort of timeline.

> but just because fusion energy will not be responsible for warming your frozen pizza next week does not mean we need to belittle a major milestone like this one. sure some damn reporter has a quota to hit and metrics to game. it just means we need to dedicate a bit more effort to read the article, read between the lines, and apply critical thinking.

When everything is presented as a "major milestone" then the term starts to lose meaning. Those with finite resources need to use imperfect heuristics to decide when to spend those resources, including the time and energy required to "read between the lines and apply critical thinking" For the majority of Gen X, "major milestone in fusion energy" has gotten the bogobit enabled by this point.

> this is the dream technology of the far future we're talking about. it doesnt even need to happen for 1000 years. if we nail it we are set for Billions of years. but obviously people are not accustomed or incentivized to think about things on that sort of timeline.

The main two issues I see with this are:

1. Maintaining a civilization capable of performing fusion research for 1000 years is an open problem

2. The estimates for how long fusion energy will last us are wildly optimistic. When you say "billions" I hear "maybe several thousand." At one point there was enough petroleum on this planet to replace all the whale-oil in use for thousands of years, and look at where we are now. Providing energy for several thousand years is great! But if we are going to think about "that sort of timeline" then fusion is a bridge-technology and with the improvements in renewables and chemical energy storage over my lifetime (which have been nothing short of astonishing!), I'm starting to suspect that it's an unneeded bridge.

I think "old and cynical" is correct, inasmuch as the human brain operates largely on pattern-matching and given that "scientists report breakthrough in X" gets proven to have a very small chance of actually affecting our lives, it becomes easy to be dismissive.

As far as "appreciate anything that might only help the next generations" I think if we are still significantly burning fossil fuels 40 years from now, the next generations are in for a lot of pain even if the very next day they get replaced by fusion, so fusion just feels like it's coming too late to make a major impact.

At the same time you are complaining about how the media handle scientific milestones (and this is unquestionably a major one), you are adopting the short term, one track, 24 hour news cycle perspective the media has spoonfed you. Global warming is only one problem. We don't need to give it a monopoly on concern. We need to push for a more nuanced, sophisticated longer term view on the major obstacles that guide a civilization. "Energy" in the public conversation should be about more than global warming.

Fusion is an energy solution that would set us up for billions of years. If we figure it out any time in the next millennium we'll have hit the jackpot.

Your take is like getting upset someone mentioned computing but didn't talk about javascript. You'd rather articles not be written about fusion until it's ready to heat your frozen pizza? God forbid we flood the news cycle with scienific progress and detract from what Elon thinks about cabbage.

Edit: for those of you feeling slightly anxious you missed new Elon gossip about cabbage, I made that up.

The problem I have with this announcement is how limiting this technology path is to the future generation of power. If you look at the details, they imploded a sphere containing hydrogen with X-rays created by the interaction of the lasers with an outer casing. As a reminder, the second stage of an hbomb is the implosion of a cylinder of hydrogen, encased by uranium, by x-rays ablating the outer case. X-Rays produced by the explosion of a A-Bomb. The lasers are thus serving as a proxy for the first stage of a H-bomb.

The critical question, how much of this design is to figure out how to produce energy from fusion with the aim of powerplant design, and how much of the design is to better replicate the second stage of a h-bomb?

Edit: I'll have to research, but I would have to expect that the announcement of the h-bomb had language that also heralded this as the beginning of a 'fusion age' of power generation.

Sketch thing is that perhaps they're try to design an hydrogen bomb that doesn't require a fission primary. Use a laser to ignite the first stage, which ignites the next and so on. What if you can get 100X boost per stage, three stages gives you a gain of 1 million.

There is just basic h-bomb efficiency gains here. By lowering the energy requirement for the fission primary by finding more efficient ways to control/generate the X-Rays you can use a smaller and smaller primary. It also occurs to me how analogous this research is to the early x-ray laser tech, that Teller claimed could shoot down warheads for the Star Wars program in the 80s. Those designs had rods that would amplify and focus the x-rays from a nuclear bomb on incoming warheads. If you could initiate the x-ray chain reaction from a laser, you would be able to launch these without breaking the 'no nuclear weapons in space' treaty.

The studies done ten years ago suggested that the best case scenario was about thirty years( one demonstration phase of 15 years and another development phase of 15 years). This was premised on the INF demonstrating ignition using indirect drive soon after operation. There was also an expectation that the INF would transistion to researching other, possibly more promising technologies like direct drive ignition.

Instead, it took over ten years to simply demonstrate ignition using the indirect drive method which was chosen specifically because it was seen as being more technologically viable. This "big breakthrough" was supposed to happen immediately after operation!

No one knows whether commercialization of inertial confinement fusion is even possible. Laser efficiency is just one small part; the work to produce and develop a system to reliably shoot hohlraum targets is an order of magnitude harder than shooting a static target in ideal conditions.

Compared to these engineering challenges demonstrating ignition is the easy part.

> Compared to these engineering challenges demonstrating ignition is the easy part.

Again, let's discuss what those challenges are. A bunch of talk about how hard it might be with no detail on what makes it hard isn't really informative to anyone.

> the work to produce and develop a system to reliably shoot hohlraum targets is an order of magnitude harder than shooting a static target in ideal conditions.

Please see my note #2 about EUV light sources and similar technology. My understanding is the timing and precision requirements are much higher here, but on the other hand the "rate" of fire would be necessarily much lower making something perhaps feasible.

From the presentation and panel today, one of the key drivers seems to be the purity of the capsules and the ability to correct for asymmetries in the implosion caused by them. The results announced today where not because of the purity of the capsule, but instead because of the correct prediction of the asymmetries in the implosion and the ability to counter that with modifications to the laser strength and timing.

The NIF isn't (as others have noted) necessarily focused on bettering these two factors.

What I'd like to understand is what are the major tradeoffs in ICF vs MCF that would drive pursuing the difficult engineering problems of one vs another.

>> My understanding is the timing and precision requirements are much higher here, but on the other hand the "rate" of fire would be necessarily much lower making something perhaps feasible.

Leaving aside rate of fire and energy efficiency you have to develop a system that can shoot the hohlraum into the reactor chamber such that the lasers can be steered onto the target with enough precision to produce ignition. I believe 100um is the miniumum for the type of targets used in the INF. This all needs to occur within the reaction chamber. Besides the thermal loads you also have residual gases within the chamber. These conditions somehow have to be accounted for or you won't get ignition.

The work they did in demonstrating ignition is great but the engineering challenges for commericalization are immense. EUV lithography took decades to develop and perfect.

>>What I'd like to understand is what are the major tradeoffs in ICF vs MCF that would drive pursuing the difficult engineering problems of one vs another.

MCF is generally seen as a much easier approach because you avoid the problems with repeatedly generating fusion conditions and you remain in the ballpark of what can be solved with better technology. You can come up with a tokamak that you can be absolutely sure will probably work with enough money; but I'm not sure the same is true with ICF.

https://www-pub.iaea.org/MTCD/Publications/PDF/TE_1704_web.p...

and

https://www-pub.iaea.org/MTCD/Publications/PDF/TE-1911_web.p...

actually layout some of the challenges. they are not as you state and there ARE tradeoffs between MCF and ICF. Most experts agree MCF is closer to realization, but that's a very well educated guess not a surity.

https://scientific-publications.ukaea.uk/wp-content/uploads/...

lays out some of the differences and tradeoffs

Even more interesting is to look at approaches which might combine the two: https://medium.com/fusion-energy-league/the-fundamental-para...

from a brief look, both publications prominently feature discussions about target survival within the reaction vessel.

Of course there are tradeoffs between MCF and ICF. They involve different sets of challenges to achieve viable fusion energy production. Neither has been demonstrated to have commercial viability to date. The consensus view that MCF is more viable is rooted in the much better understanding of how to comercialize a MCF reactor.

The issue is the NIF isn't trying to study power generation, their mission is determining the health of our weapons stockpile. This experiment is a lab where they can test different casings and different hydrogen mixtures and this probably feeds real-world data into their supercomputer models. All this to determine how a 50 year old h-bomb would operate now. They also can find out how they can improve them without being able to go to Nevada and detonate one underground.

> The US$22-billion ITER project — a collaboration between China, the European Union, India, Japan, Korea, Russia and the United States — aims to achieve self-sustaining fusion, meaning that the energy from fusion produces more fusion

It's refreshing to see collaboration like this still happening, even with all the tensions between all those countries.

It was happening for decades. Unclear how the war affects it now, but I believe most RU hardware (cryo and magnets) was already delivered/installed.

It is not a nuclear-fusion lab. It is a nuclear-weapons lab. What it means is they got a better system to model weapons.

What it means to DoE higher-ups is that if they paint it as a breakthrough in energy production, they can unlock funding increases. So they do.

Its so frustrating to see so many people not understand this means relatively nothing towards the goal of power generation. We'll see how many details are released to the public and other scientists and how much is kept top-secret. They very well could have achieved this 10 years ago using similar materials as in the h-bomb, but kept that secret.

Never thought I'd witness a resonance cascade... Let alone, create one.

I read a comment here before saying that the purpose of the NIF is to get more data for making hydrogen bombs, while being unable to test actual bombs due to treaties. And that the press releases they have about fusion energy are mainly bullshit for the press.

Is there anything to that claim or is it bullshit?

Part of their job is to make sure that the nuclear stockpile is still reliable. Since they can't actively blow up a nuke nowadays, they instead rely on experiments and simulation.

However, like a lot of the Department of Energy's work, their nuclear research also relates heavily to civilian applications. They are not 'bullshit' for the press.

If I am not mistaken, there is also no current technology to actually harvest that energy, right?

So the lasers need at least 10x the energy they themselves put into the target. Conversion is still unclear. Are we even at 1% to a demo reactor, let alone a commercially viable one?

"To demonstrate that the type of fusion studied at NIF can be a viable way of producing energy, the efficiency of the yield — the energy released compared to the energy that goes into producing the laser pulses — needs to grow by at least two orders of magnitude."

Lots of claims on HN that we were one order of magnitude away. Similar claims were made in other threads made about how we were really close to commercialization with better lasers. This article has the right take, this research is valuable but its impossible to say anything about whether indirect intertial confinement fusion is commercially viable. Best case scenario [1], we still have a way to go.

[1]: An Assessment of the Prospects for Inertial Fusion Energy (2013)

This article was a welcome relief after many of the others the past couple of days. Well-written, clear, optimistic but not hand-wavy, and certainly not "breathless" or loaded with PR.

“However, while the fusion reactions may have produced more than 3 megajoules of energy — more than was delivered to the target — NIF’s 192 lasers consumed 322 megajoules of energy in the process.”

Well that says it all right there and really cuts through all the fluff on this story. Sounds like a nice milestone that has absolutely nothing to do with our ability to get a useful amount of energy back out. We are still at least 2 orders of magnitude away from a power source.

Do you have a background in this field?

I do not. I would say if I did.

To be blunt then: your comment does not add to the discussion in a substantive way.

And plainly, if we all needed to be nuclear physicists to talk and think about this, there would be no comments and no articles. Your comment also does not advance the conversation.

The next steps in the story are, how easy is it to make this process more efficient? Is there something fundamental that is lessening the efficiency, or is it a matter of needing to buy part X or laser X or variant y to improve efficiency?

Most research achievements are wildly inefficient because efficiency is not really their goal, they're looking for proof of concept. Once the concept is proven, you step into a whole different ballgame when efficiency is the goal. (Hopefully, who knows.)

There's a reason for optimism as for the lasers' efficiency. But there are many other aspects that were only approached theoretically so far, such as how to get useful energy out of the reactor.

This is closer to pure science. This is in a lab. They aren't even attempting to make a "power source" so much as testing and verifying the theory.

And that is great! My issue is the way the science is reported in the media making it sound like we are on the cusp on some kind of energy revolution, when we very much are not and don’t even have a line of sight on it yet. But then I suppose public hype will lead to Congress dollars and nothing gets done without those.

Seemed like this article in Nature did pretty good. They explicitly said: “I don't want to give you a sense that we're going to plug the NIF into the grid: that is definitely not how this works,” and “Although positive news, this result is still a long way from the actual energy gain required for the production of electricity,” said Tony Roulstone, a nuclear-energy researcher at the University of Cambridge, UK, in a statement to the Science Media Centre.

Still, “the NIF experiments focused on fusion energy absolutely are valuable on the path to commercial fusion power”, says Anne White, a plasma physicist at the Massachusetts Institute of Technology in Cambridge.

There is a pretty solid disconnect between the public's perception of how science works and happens and how it does. For this to really matter, someone else needs to duplicate the experiment and verify the data. It's a huge win if the hype can lead to that.

To me this is the most important part of the article,

“It’s a big milestone, but NIF is not a fusion-energy device,” says Dave Hammer, a nuclear engineer at Cornell University in Ithaca, New York.

Herrmann acknowledges as much, saying that there are many steps on the path to laser fusion energy. “NIF was not designed to be efficient,” he says. “It was designed to be the biggest laser we could possibly build to give us the data we need for the [nuclear] stockpile research programme.”

No one at this lab is saying it is ready for commercial sales. The technical Q&A even featured them saying high energy lasers might not be the best tech for driving the fusion in a commercial setting- https://youtu.be/Eke5PawU7rE?t=5250

We already have lasers which are an order of magnitude more efficient than the one the lab was using. This is about a proof of concept, not about being an actual power plant.

Also, as I understand it, the whole thing scales up very well. Now that we can "ignite" fusion in the little pellet used for testing, we can scale up to ignite a larger amount of fuel in order to get out much more energy.

They put 2.05 MJ in and got 3.15 MJ out. The 300MJ comes from equipment inefficiency- remember this is a research facility to prove plasma physics and ignition itself, mainly designed for weapons research. The fact they got definite fusion at 1.53x input energy is not just breakeven, but net gain. This is the first time ever we've gotten net gain, in any fusion device. A lot of people in gov leadership thought fusion was impossible, the fact they can said they did it, despite 1980s era technology and a shoestring budget for decades, really says something.

ICF is just one fusion technology, and probably not the best one, but fusion in general will probably see a massive budget increase worldwide now that it's been proven it can be done.

I fully expect a working fusion plant of some kind by 2030, assuming funding increases. Once we get them commercialized; coal, wind, solar and other power production will be obsolete for the most part. We can also use fusion heat to separate waste into it's base elements (you can recycle anything!), and help make any process needing a lot of thermal or electrical energy more efficient.

How much easier to build and safer are they than fission plants? Otherwise they will still be politically bottlenecked.

IANAScientist, but I believe I read that fusion requires a steady input of energy to continue the reaction. So much, much safer than fission, which is a chain reaction that can go out of control at any time, fusion, you just turn off the power and it stops.

IIRC

Not useful for commercial energy production, but useful if you want to study self-propagating fusion explosions in a bomb, which is the actual purpose of the facility.

NIF is really confusing the issue (I think this is the first pop-sci article Ive seen that even mentions nuclear weapons, and its relegated to one sentence surrounded by discussion of commercial energy). They're (justifiably) excited because after a decade of trying, they've finally managed to fulfill the facilities primary purpose and recreated an important aspect of the conditions inside a thermonuclear bomb.

But they're obviously reluctant to foreground the bomb stuff, so instead they hand-wave in the direction of energy production, which doesnt make a lot of sense, and leads to a lot of folks pointing out that the energy needed to get the reaction going is more then just the laser pulse itself, which would make a big difference for commercial production, but doesn't really matter for the experiment theyre trying to perform

This means we might have power as cheap and plentiful as solar and lithium batteries are today in a couple of decades, when solar and Batteries will be even cheaper.

That's a little optimistic.

Or do you mean in the arctic or something?

In more human terms :

    They fused in 0.6 kwh energy.
    Which produced 0.9 kwh energy.

These megajouls are confusing for me. kwh is what we use for resident home power billing unit.

Moreso: they managed to put in 0.6 kWh of the highest grade of energy imaginable, and got out 0.9 kWh of hot neutrons, the lowest grade of energy imaginable (short of neutrinos).

This is quite frankly awesome.

Now if it's sustainable, we could definitely use the power to do desalination plants here on the west coast, since it's a super energy hungry process. And the hydrogen/deuterium could come from the salt water its processing. Then, it's just a matter of what to do with the salt generated.

https://backreaction.blogspot.com/2021/10/how-close-is-nucle...

Informative and sobering. That's what we need, not press conferences.

I don't know. This is not "we're there", not by a long shot. But it still seems to me to be worth throwing a lot of confetti over.

More discussions: https://news.ycombinator.com/item?id=33971377

It means practical fusion power, which has been 20 years away for the last 60 years, will now be just 10 years away for the next 30 years.

Followed, of course, by being 5 years away for the next hundred years.

Big oil is much more competent adversary than science.

Well not even that, there isn't even a roadmap to improve laser efficiency 100x

Even a 90% efficient high powered laser is way beyond fusion in terms of technical difficulty.

But am I wrong in thinking that it’s conceivable that we could see a further doubling in Q-plasma? If that happens, perhaps along with significant-but-plausible improvement in laser efficiency, then things could start to look fairly promising.

Did you reply to the wrong comment?

Nope! My point was that your comment assumes no further increase in Q-plasma. But the experience of the last few years suggest that there’s reason to hope that won’t be the case. We can come at the problem from both sides.

Didn't someone the same thing about atomic power back in the middle ages?

But that's OK, fusion is a hard thing to do outside of a star. It may take decades longer for it to be practical but doesn't the potential return on investment make it worthwhile to keep working on it now?

I just want to say, I thought this was a very funny joke by me.

Call me a cynic but the only meaningful thing out of this is that they got their funding secured.

Can someone calculate how much installed solar power could we get by pausing fusion r&d for the next 10-20 years?

Not enough.

About zero if China refuses to sell us solar panels due to some conflict.

It’s really stretching things to say it produced more energy than went into the reaction.

Energy in:energy out is a ratio of the amount of energy that goes in, not the portion of the energy that eventually hits the target. It’s like measuring car engine efficiency by measuring how much of the energy reaching the wheels translates into motion - and ignoring the massive losses that lead up to that point.

Using your analogy, this experiment was to see if more energy could be imparted as motion then was put into the tires from the engine, due to a mechanism of physics of the tires that would release energy. Many people doubted our ability to exploit this mechanism of physics.

Now that our ability to exploit this mechanism of physics has been demonstrated, development of power sources that exploit this mechanism can begin to receive funding.

Reply on news.ycombinator.com